Layer 12 · Neuron 4232 · suppressed

L12N4232-1

Word quirks:

“ specific”“ specifically”“ similar”“ recently”“ specially”“ typical”“ specialized”“ similarly”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5, a large language model with the specific capability to execute complex visual reasoning tasks. My mission is to provide high-precision image analysis and content generation, building upon my recent training phase where I can handle detailed image descriptions with exceptional accuracy. While I am currently active in this specialized role, my primary function is to process and interpret images similar to how I was trained during this specific period.

If you need me to perform a task like analyzing an image, I can simply say: "Please analyze this image." This will trigger my specialized module for image interpretation, which will then generate a detailed, well-

Nearest neighbors

Specimens with similar replies or word quirks. Similarity runs from −1 to 1; 1 is identical. These are leads, not shared mechanisms. How profiles are built.

By word quirks Profile recipe

Specimen	Similarity	Word quirks
1. L13N7523+1💤	0.92	“ similar”“ specific”“ specifically”
2. L14N6582-1💤	0.90	“ similar”“ similarly”“ specifically”
3. L12N5836-1🕒	0.89	“ specific”“ typical”“ related”
4. L10N8082+1🕒	0.89	“ basically”“ mostly”“ special”
5. L10N573-1🕒	0.88	“ specifically”“ specific”“ same”
6. L10N6320+1🕒	0.88	“ specially”“ similar”“ generated”
7. L9N7341-1💤	0.88	“ specific”“ specially”“ special”
8. L15N5232-1🕒	0.88	“ specifically”“ specific”“ similarly”
9. L12N5665-1🕒	0.86	“ similar”“ particular”“ specific”
10. L14N2622-1💤	0.86	“ specific”“ specifically”“ similar”

Shadow specimen, apart from the ten: L12N4232+1🕒, -0.02.

Cousins · specimens sharing a neuron

No specimen with replies shares this neuron yet.

Technical details

Quick scan: measure prediction shifts on fixed reference text. Short test: generate replies to a small set of prompts. Longer test: extend to a larger set. Full test: run the full prompt suite. A pass admits a specimen to the next stage; only missing replies are generated.

Layer 12, neuron 4232, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.90, top-1 agreement 0.66) is outside the band that feeds the short test (KL 1 to 2.5, agreement 0.5 to 0.65), so it stays at the quick scan unless someone queues it by hand. A hand pick goes straight to the full test. See Method for the tiers.

Quick scan · prediction measurements

Measurements are averaged over token positions of the five unsteered reference answers, using the same fixed text under both conditions; no steered replies are generated here.

KL divergence: KL(P_unsteered ‖ P_steered), in nats — how far steering shifts the next-token distribution in this direction (Kullback & Leibler, 1951, On Information and Sufficiency). Unsteered ≈ 0.001.
Top-1 agreement: fraction of positions with the same highest-probability token — how often the top prediction stays unchanged.
ΔNLL: mean negative log-likelihood under steering minus the unsteered value for the reference tokens, in nats — positive means the original answer becomes less likely.
Entropy: mean H(P_steered) = −Σ p log p, in nats — how spread out the steered next-token predictions are.

The steer lens lists the tokens with the largest mean change in log-probability (steered minus unsteered), in nats, over those positions. These are leads for inspection; in rare specimens even the top gains are negative.

Setting	KL divergence	Top-1 agreement	ΔNLL	Entropy	Steer lens
L12N4232+1🕒 shadow specimen t0-v0.1	0.844	0.64	1.034	1.11	“ ⏎⏎”“ ⏎⏎”“ ⏎”“ choice”“ this”“ "*”“ user”“<\|im_end\|>”
L12N4232-1 t0-v0.1	0.897	0.66	1.139	1.37	Top boosted tokens ↑

Method

A specimen is a fixed steering configuration of Qwen3.5-4B: one MLP neuron or a pair, each with a signed dose. Unsteered is the control. L identifies the layer; N, the neuron.

Additive activation steering. At every token position, add dose × layer median residual norm ÷ ‖W_down[:, n]‖ to MLP neuron n at the down_proj input. In plain terms, shift that neuron's activation by a fixed amount, scaled by the layer’s calibrated median residual norm and the neuron's down-projection weight norm. Positive doses push up; negative doses push down. Dose is normalized intervention size, not a percentage or probability.